Download 3D models from any website.
GLB, glTF and the rest, including models that three.js sites assemble at runtime. Rendered previews before you download, no extension to install.
01Why 3D is the hardest asset to take
A WebGL site does not hand you a model file. It fetches geometry, textures and animations over the network, sometimes in pieces, sometimes as JSON that only three.js understands, and assembles the scene in memory. View-source shows you nothing; there is no right-click for a spinning product model.
pagehaul’s deep scan runs the page in a real browser and records what it actually fetches. GLB and glTF files are collected whole, runtime-assembled models are recognised from the network traffic, and each model gets a rendered preview in the 3D tab so you know what you are taking before you take it.
02Formats, previews, and the rest of the page
GLB and glTF are the web’s native 3D formats, and Draco compression is handled for previews. OBJ, FBX and STL are collected where pages serve them directly. And because a scan reads the whole page, the textures, environment images and audio a scene uses are sitting in the neighbouring tabs, along with every image and every SVG.
03Working on a client site?
The same engine that reads every asset on a page will also tear the page down: type scale, colour contrast against WCAG, spacing grid, speed, mobile and conversion, written up as a report you can put your name on and send. Useful when the job is not the files but the argument for a rebuild.
04Questions people ask
How do I download a 3D model from a website?
Paste the address into pagehaul and run a deep scan. Every model the page loads appears under the 3D tab with a rendered preview, ready to download.
What 3D formats does it find?
GLB and glTF primarily, the formats the web actually ships, plus OBJ, FBX and STL where pages serve them. Draco-compressed models preview too.
Can it extract models from three.js websites?
Yes. Sites built on three.js often assemble models at runtime instead of loading one file; a deep scan watches the page's network like a browser does and records the model files it fetches.
Do I need a browser extension?
No. pagehaul runs entirely in the browser you already have; paste a link and scan.
Can I use a downloaded 3D model in my own project?
Downloading gives you no rights to it. Models carry licences like any asset. Built for your own sites, migrations, backups and reference.
Paste a link and see every file the page is built from.
Scan a page free